Research article Received: 31 October 2014,
Revised: 24 February 2015,
Accepted: 30 March 2015
Published online in Wiley Online Library: 5 May 2015
(wileyonlinelibrary.com) DOI 10.1002/bmc.3482
Ion-exchange vs reversed-phase chromatography for separation and determination of basic psychotropic drugs Anna Petruczynik*, Karol Wróblewski, Michał Deja and Monika Waksmundzka-Hajnos ABSTRACT: Ion exchange chromatography, an alternative to reversed-phase (RP) chromatography, is described in this paper. We aimed to obtain optimal conditions for the separation of basic drugs because silica-based RP stationary phases show silanol effect and make the analysis of basic analytes hardly possible. The retention, separation selectivity, symmetry of peaks and system efficiency were examined in different eluent systems containing different types of buffers at acidic pH and with the addition of organic modifiers: methanol and acetonitrile. The obtained results reveal a large influence of the salt cation used for buffer preparation and the type of organic modifier on the retention behavior of the analytes. These results were also compared with those obtained on an XBridge C18 column. The obtained results demonstrated that SCX stationary phases can be successfully used as alternatives to C18 stationary phases in the separation of basic compounds. The most selective and efficient chromatographic systems were applied for the quantification of some psychotropic drugs in fortified human serum samples. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: IEC-HPLC; RP-HPLC; psychotropic drugs; optimization; quantitative analysis
Introduction
1700
Depression is one of the most frequently occurring psychiatric conditions affecting the economic and social functioning of people all over the world. Therapeutic drug monitoring of antidepressants provides the possibility of reducing side effects and optimizing the treatment of patients with depression. For this reason, there is a need to develop analytical methods for the analysis of these drugs, especially for their determination in biological fluids (Plenis and Bączek, 2011). A wide variety of analytical techniques has been used to determine different psychotropic drugs. Nowadays liquid chromatography in reversed-phase system (RP HPLC) is the most frequently used method (Nikitas and Pappa-Louisi, 2009). However, the separation of basic compounds in LC-RP systems still remains problematic owing to their interactions with free-residue silanol groups. Organic bases appear in aqueous solution as ionized and unionized forms that interact differently with the stationary phase surface-active sites (ion-exchange, hydrophobic and H-bond interaction; McCalley, 2010). The peak shape of basic compounds depends on the kinetics of the different interactions. In aqueous mobile phases, interactions of ions are usually stronger and slower than hydrophobic interactions. This situation leads to peak tailing, low efficiency and poor column-to-column reproducibility (Newby et al., 2011). Interactions with the silanols can be reduced by the use of mobile phases buffered at low pH, when silanol ionization is suppressed, or at high pH, to suppress solute ionization, addition of anionic ion-pair reagents to form neutral associates, addition of organic amines as silanol blockers, or the use of an appriopriate stationary phase (Nawrocki, 1997). Although advances in column technology, especially in the use of high-purity silica, have given considerable improvements for the analysis of bases, low
Biomed. Chromatogr. 2015; 29: 1700–1707
efficiency and tailing peaks continue to be problematic for these compounds under some circumstances. Therefore, an alternative approach to modulate the retention, peak shape, systems efficiency and separation selectivity of ionizable compounds using of ion exchange chromatography (IEC) is required (Long et al., 2012). The application of IEC allows high efficiency, symmetrical peaks and usually good separation to be achieved using relatively simple eluents containing only buffer or a mixture of buffer and organic modifier (Long et al., 2013b). Additionally, mobile phases applied in IEC contain usually low concentrations of organic modifiers (or do not contain any), which makes the method environmentally friendly. In eluent systems at constant pH, the retention of basic compounds in the chromatographic mode is inversely proportional to ionic strength and (if addition of organic modifier is used) to kind and concentration of organic modifier in mobile phases (Croes et al., 1995). Because of the similarity of ion-exchange and ion-pair HPLC retention, many separations that are possible using IEC can also be achieved using ion-pairing chromatography (IPC). For the separation of typical small-molecule samples, IPC may have certain advantages: higher column efficiencies, easier control over selectivity and resolution, and more stable and reproducible columns (Snyder et al., 1997). However, both IEC and reversed-phase
* Correspondence to: A. Petruczynik, Department of Inorganic Chemistry, Medical University of Lublin, Chodzki 4a, Lublin 20-093, Poland. Email: [email protected] Department of Inorganic Chemistry, Medical University of Lublin, Chodzki 4a, Lublin 20-093, Poland Abbreviations used: DEA, diethylamine; IEC, ion exchange chromatography.
Copyright © 2015 John Wiley & Sons, Ltd.
Ion-exchange chromatography for analysis of psychotropic drugs
Biomed. Chromatogr. 2015; 29: 1700–1707
Mixed mode C18/SCX phases also give good peak shape for ionized bases, e.g. propranolol and furosemide were determined in human plasma (Walshe et al., 1996). The analysis of drugs in biological samples such as plasma, urine and saliva requires robust, precise and fast methods. The success of such analyses is largely dependent not only on the choice of the most optimal chromatographic system but also on the quality of the sample preparation. Different SCX stationary phases for SPE were also used for preparation of biological samples containing basic analytes (Brown et al., 2008; Gilart et al., 2014). The aim of this paper was to investigate the various parameters influencing the retention, peak shape, efficiency and separation selectivity of some psychotropic drugs on an SCX column. The effect of the use of MeOH or MeCN as organic modifiers, ionic strength and type of buffer applied in the mobile phase was examined. The results obtained on the SCX column were compared with those obtained on C18 stationary phase. Systems with the best parameters and highest selectivity were applied for the quantification of some psychotropic drugs in fortified samples of human plasma.
Experimental The analysis was performed using liquid Shimadzu chromatograph LC-10 ATVP equipped with a Luna SCX 150 × 4.6 mm, 5 μm particle size column (Phenomenex, USA) or an XBridge C18 column from Waters (150 × 4.6 mm, 5 μm). The properties of the columns are presented in Table 1. The chromatograph was equipped with a Shimadzu detector SPD–10 AVVP and a Rheodyne 20 μL injector. The detection was carried out at 254 nm wavelength. All chromatographic measurements were carried out at 22°C controlled by a CTO-10ASVP thermostat. Eluent flow rate was 1.0 mL/min. The column pressure during the analysis was in the range of 250–300 psi. MeCN and MeOH of chromatographic quality were from E. Merck (Darmstadt, Germany) and diethylamine (DEA) (analytical reagent grade) was from Sigma-Aldrich. Potassium dihydrogen phosphate, sodium dihydrogen phosphate, calcium dihydrogen phosphate, ammonium dihydrogen phosphate, phosphoric acid (85%), sodium chloride, potassium chloride, ammonium chloride, cesium chloride, lithium chloride, barium chloride, calcium chloride, ammonium acetate and acetic acid were of p.a. quality (Polish Reagents Gliwice, Poland). The pH of buffers used in experiments was measured using a pH meter CP-505 (Elmetron, Zabrze, Poland) in aqueous solutions. Analysis of serum samples was performed using a liquid chromatograph LaChrom Elite (Merck) equipped with autosampler diode array detector (DAD). All chromatographic measurements were carried out at 22°C with an eluent flow rate of 1.0 mL/min using the same IEC column. The DAD was set in the 200–400 nm range and quantitative analysis was performed at 238 nm for all compounds. The chromatographic data was acquired and processed by LaChrom Elite HPLC software (Merck). All chromatographic parameters such as retention times (tR), asymmetry factor (As) (calculated as 10% of peak height) and theoretical plate number (N/m) were calculated by software CLASS-VP 5.0 controlling the chromatograph.
Table 1. Physicochemical properties of columns Column
Particle Pore Surface size (μm) size (Å) area (m2/g)
Xbridge C18 SCX
Copyright © 2015 John Wiley & Sons, Ltd.
5 5
130 100
Carbon load
pH limits
185 18% 1–12 400 0.55% sulfur load 2–7
wileyonlinelibrary.com/journal/bmc
1701
IPC have their advantages and disadvantages. One of the main advantages of IEC is that there is only one interaction involved in the separation: the analytical species interact with the stationary phase whereas in IPC two interactions are involved with the separation – the ion-pairing reagent interacting with stationary phase of the column and the species interacting with the ion-pairing reagent. As a result, IEC may have more matrix tolerance. The disadvantage of IEC is that these columns typically are much more expensive than reversed-phase columns and ion-exchange columns may not always be well established, which could lead to poor reproducibility from column to column (Neubauer, 2009). The effects of changes in eluent composition, kind of buffer, ionic strength and pH on the retention of different basic compounds on C18/strong cation-exchange (SCX) columns in eluent systems containing acetonitrile or methanol and sodium, potassium or ammonium phosphate in different concentrations and at different pH values have been investigated by some authors (Walshe et al., 1995). An SCX column was applied to the analysis of various basic drugs, e.g. β-blockers, using a mobile phase containing methanol (MeOH) and ammonium formate with trifluoroacetic acid at pH 2.45 (Law and Appleby, 1996), trazodone in a mobile phase system containing acetonitrile (MeCN) and ammonium phosphate adjusted to pH 6.0 (Li-Boa et al., 2014), MeOH with sodium phosphate buffer adjusted to pH 4.8 with triethylamine (Ghosheh et al., 2000), MeOH and ammonium perchlorate (Morgan et al., 2003); often mixtures of MeCN and ammonium formate or acetate as additives to mobile phase were used ( Jiang et al., 2011; Koseki et al., 2005). Ephedrine alkaloids in dietary supplements were determined on SCX column with mobile phase containing MeCN, sodium phosphate buffer at pH 3.0 (Niemann and Gay, 2003) and alkaloids from hallucinogenic mushrooms with mobile phase containing EtOH and phosphate buffer with the addition of NaCl (Laussmann and Meier-Giebing, 2010). Ion-exchange chromatography has been used successfully to separate different basic compounds in biological samples, e.g. alkaloids in urine (Schonberg et al., 2006), or in plasma (Ghosheh et al., 2000), some basic drugs in plasma (Law and Appleby, 1998), the antidepressant drug trazodone in human plasma ( Jiang et al., 2011), mitrazapine and its metabolite in plasma (Morgan et al., 2003) and the antihyperglycemic drug metformin in human plasma (Koseki et al., 2005). Some tricyclic antidepressants were determined in plasma samples on methylcellulose-immobilized restricted access media column with strong cation-exchange groups on an internal surface (MC-SCX; Kawano et al., 2005). Long et al. (2012) described the results of the influence of salt concentration and buffer pH on the peak shape obtained for some basic compounds under high sample loading conditions. Basic compounds were effectively separated on an SCX column and only a small increase in peak width was observed as the loading amount of analytes increased, which can be important for preparative separation of these substances. The influence of type of buffer at different values of pH, ionic strength, type and concentration of organic modifier on the retention of selected alkaloids on SCX column was also investigated (Petruczynik and Waksmundzka-Hajnos, 2013). The SCX column is also used coupled with an RP column for multidimensional separation owing to the good orthogonality between SCX and RP separation mechanisms with similar mobile phases. For example, a 2D-LC system, SCX/RP, has been developed to eliminate nonalkaloids and to achieve symmetrical peak shape and high orthogonality selectivity for alkaloid separation (Long et al., 2013a).
A. Petruczynik et al. Extraction procedure Solid-phase extraction (SPE) was carried out using Bakerbond SPE C18 endcaped columns and an SPE chamber–Baker SPE–12G ( J. T. Baker, Philipsburg, PA, USA). The SPE method was optimized and the best procedure in terms of recovery and purification was selected for sample preparation. A 1 mL aliquot of human serum sample was fortified by escitalopram, sulpiride and zolpidem at concentration 0.5 μg/mL and was incubated at 37°C for 60 min. A C18 endcaped cartridge was preactivated by passing through 3 mL of MeCN, 10 mL of MEOH and 30 mL of water. A 1 mL aliquot of serum was mixed with 16 mL of water and 3 mL of 1 M sodium carbonate. The sample preparation was mixed and loaded onto the C18 cartridge at a flow-rate of 2 mL/min. The cartridge was washed twice with 5 mL of water. Finally, the drugs were eluted with 5 mL (two fractions, 3 + 2 mL) of acetonitrile and HCOOH (80%; 19:1). The sample was evaporated to dryness, the residue was dissolved in 0.5 mL of acetonitrile and a volume of 20 μL was injected into chromatographic systems.
Validation protocol The proposed method was validated by linearity, limit of detection (LOD), limit of quantification (LOQ) and precision . Method linearity was studied by analyzing solvent-based standard solutions in triplicate at eight concentrations ranging from 0.1 to 10 μg/mL for escitalopram, sulpiride and zolpidem. The LOD and LOQ were calculated according to the formulas LOD = 3.3(SD/S) and LOQ = 10(SD/S), respectively, where SD is the standard deviation of the response and S is the slope of the calibration curve. aThe accuracy of the method was tested by performing recovery studies. The average recovery was 98.6% for escitalopram, 89.66% for zolpidem and 71.86% for sulpiride.
Results and discussion Psychotropic drug standards (Table 2) were chromatographed on an SCX column in eluent systems containing different buffers at pH 2.5 consisting of various chlorides and organic modifiers or
on a C18 column in eluent systems containing acetonitrile, phosphate buffer at pH 3.5 and 0.025 M DEA. These chromatographic systems were compared in terms of retention of investigated drugs, differences in selectivity, peak shape and performance. The use of different buffers in mobile phase – acetate, formate and phosphate without addition of chlorides – were tested, but asymmetrical peaks and low systems efficiency were obtained.
Selection of the organic modifier The first part of our work was a study of the retention behavior of investigated drugs on an SCX column in eluent systems containing phosphate buffer at pH 2.5, KCl and MeOH or MeCN as organic modifiers. Great differences in separation selectivity were observed between systems with mobile phases containing different organic modifiers (Table 2). The retention of psychotropic drugs was stronger in mobile phase with MeOH. Also, differences in peak shape and system efficiency were obtained in eluent systems containing MeOH or MeCN. Symmetry of peaks was significantly improved in mobile phase containing MeCN. In systems with MeOH as organic modifier, only in one out of 16 psychotropic drugs was the value of As in the acceptable range (0.8–1.5), while in systems with mobile phase containing MeCN, for 15 drugs the symmetry of peaks was acceptable and for 12 of them As values were excelent (0.9 < As < 1.2). The addition of MeCN to mobile phases also caused an increase in systems efficiency. In systems containing MeOH, N/m values were >25,000 for five out of 16 psychotropic drugs, and in systems containing MeCN as organic modifier in the mobile phase, for 15 investigated drugs, the number of theoretical plates (N/m) was >25,000. Because of the low efficiency and high peak assymmetry, the resolution of investigated drugs in systems with methanol got worse and much better results were observed when acetonitrile as modifier was applied. For this reason in further experiments MeCN was used as organic modifier.
Table 2. Values of retention time (tR), asymmetry factor (As) and theoretical plate number (N/m) for investigated psychotropic drugs obtained on an SCX column in eluent systems containing 25% organic modifier, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl Investigated psychotropic drugs
Abbreviation
Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moclobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
R M E T Sd I Ro A Z H Mo B F Se Fl S
MeOH
MeCN
tR
As
N/m
tR
As
N/m
17.44 31.84 22.21 10.78 65.13 26.83 19.43 * 51.61 81.27 15.33 15.67 16.03 22.21 12.43 20.83
3.63 4.56 2.89 1.55 4.60 3.55 2.85
18,470 9910 23,070 31,660 40,030 14,230 22,160
4.21 2.74 2.30 1.94 1.42 2.99 2.00 1.54
8870 10,890 30,190 38,500 35,720 14,960 14,620 33,130
8.18 11.03 7.47 6.21 10.23 8.21 8.50 4.30 13.22 19.70 8.39 8.34 8.34 7.54 5.37 9.78
1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
41,590 41,350 41,580 43,440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
1702
* Fuzzy peak.
wileyonlinelibrary.com/journal/bmc
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs Selection of buffer cation The next stage of the experiments concerned the effect of different cations in salts used to prepare the phosphate buffers as mobile phase additives. The use of different salts caused differences in retention, separation selectivity, system efficiency and peak symmetry. The selectivity and sequence of elution for the psychotropic drugs in the systems containing buffers with various cations were different (Table 3). For most drugs better selectivity was obtained in systems containing sodium salts. Also, differences in peak shape and system efficiency were obtained in eluent systems containing salts with different cations. The worst peak shapes were observed in eluent containing sodium salts. Only nine out of 16 investigated drugs had As values in the acceptable range, whereas in the system with calcium cations As values for 12 compounds were acceptable. The most symmetrical peaks were obtained in systems containing potassium salts – for 15 psychotropic drugs As values were acceptable and for 12 they were excellent. High efficiency was also obtained in systems containing potassium salts in the mobile phase: for 15 drugs, N/m was >25,000, while in systems containing calcium or sodium salts for 9 and 12 compounds N/m was >25,000, respectively. For this reason further investigation was performed in eluent systems containing potassium phosphate buffer.
Selection of the chloride salt cation Because the retention of the investigated psychotropic drugs significantly depends on the type of cation used to prepare buffer solutions, mobile phases containing potassium phosphate buffer and different chlorides were used in the next set of experiments. The retention and separation selectivity obtained using buffers containing chlorides with various cations (pH and the other eluent components are the same) were significantly different (Table 4).
The eluent strength decreased from the eluent containing barium chloride to the eluent containing lithium chloride according to Hofmeister series for most investigated drugs. Displacer strength in cation-exchange chromatography highly depends on the valency of cations. Thus the displacer strength of bivalent cations is higher than that of monovalent and trivalent ones (Snyder et al., 1997). Because of that, in our investigations analysed compounds were eluted from the column after a very short time and they were poorly separated in systems containing chlorides of bivalent cations (calcium and barium). In mobile phase containing cesium chloride, separation of most drugs was also poor. The similar retention times and good separation selectivities were obtained in eluents containing additions of ammonium, potassium and sodium chlorides. Various chlorides cause also differences in peak symmetry, e.g. for alprazolam As = 2.66 in the system with CaCl2, whereas in the system with LiCl, As = 1.19; for flupentixole As values were 0.98 in systems with CsCl and KCl and 1.00 in systems with added LiCl, but in the system with BaCl2 the peak was very asymmetrical. However, for most investigated psychotropic drugs in systems with potassium phosphate and different chlorides, peaks were symmetrical; only in a few cases As values dif not posess the optimum range. The differences were also observed for system efficiency, for example, for rivastigmine N/m was 18,390 for the system with CaCl2 and 41,590 for the system with KCl; for flupentixol N/m was about 19,000 for the system with BaCl2 and N/m was >40,000 for systems with NH4Cl, KCl, NaCl and LiCl. For nearly all investigated drugs, worse system efficiency was obtained in eluents containing chlorides with bivalent cations compared with eluent systems with monovalent cations. The most efficient system for most drugs was the system with addition of KCl to mobile phase – for 15 from 16 investigated drugs N/m was >25,000.
Table 3. Values of tR As and N/m for investigated psychotropic drugs obtained on an SCX column in eluent systems: (1) 25% MeCN, buffer at pH 2.5 [100 mM Ca(H2PO4)2, 100 mM H3PO4] and 100 mM CaCl2; (2) 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl; and (3) 25% MeCN, buffer at pH 2.5 (100 mM NaH2PO4, 100 mM H3PO4) and 100 mM NaCl Investigated 25% MeOH 100 mM CaCl2, 100 mM Ca (H2PO4)2, H3PO4, pH 2.5 psychotropic drugs As N/m tR Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moclobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
5.3 6.18 4.26 1.88 5.76 4.66 5.09 3.63 8.06 * 4.47 4.76 4.77 4.13 3.20 5.44
1.27 1.39 1.00 2.47 1.08 1.06 1.11 2.32 1.74
26,680 27,230 27,950 13,570 24,330 28,930 28,570 5530 21,970
1.05 1.00 0.84 1.08 1.04 0.97
30,640 30,970 30,460 20,010 22,300 30,110
25% MeOH 100 mM KCl, 100 mM KH2PO4, H3PO4, pH 2.5 tR 8.18 11.03 7.47 6.21 10.23 8.21 8.50 4.30 13.22 19.70 8.39 8.34 8.34 7.54 5.37 9.78
As 1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
N/m 41,590 41,350 41,580 43,440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
25% MeOH 100 mM NaCl, 100 mM NaH2PO4, H3PO4, pH 2.5 tR
As
13.87 12.04 7.36 9.69 14.48 21.78 9.83 * 24.58 * 13.05 8.85 12.61 6.69 6.82 18.82
N/m
2.60 1.46 0.91 1.01 0.75 2.03 1.09
19,730 39,220 38,240 37,520 18,380 31,130 38,590
2.07
27,350
1.80 0.93 1.12 0.97 0.90 1.15
39,310 44,240 28,960 30,640 34,520 39,270
1703
* Fuzzy peak.
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
A. Petruczynik et al. Table 4. Values of As and N/m for investigated psychotropic drugs obtained on an SCX column in eluent systems containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM of different chlorides Investigated psychotropic drugs
CaCl2
As
N/m
As
Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moklobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
0.9 1.27 1.04 1.04 1.00 1.05 1.09 2.12 1.47 1.02 1.00 * 0.67 0.90 1.01 1.03
26,570 27,210 26,750 23,510 24,840 28,930 28,720 10,980 24,120 23,480 20,770
1.02 1.27 1.00 1.01 1.02 1.04 1.04 2.66 1.49 1.04 0.98 * 0.84 0.98 1.02 1.03
BaCl2
19,290 31,934 19,560 25,820
CsCl
N/m 18390 28600 28190 28040 24,520 30,340 29,550 7250 23,810 21,500 21,000 29,700 32,770 21,500 25,180
NH4Cl
As
N/m
As
1.06 1.22 0.92 0.98 0.95 0.96 0.99 1.47 1.34 0.94 0.96 * 0.98 0.89 0.99 1.01
33530 35150 33450 33060 28,990 35,830 36,700 11,760 30,750 29,560 27,470
1.11 1.34 0.91 0.96 0.95 0.98 1.00 1.30 1.52 1.26 1.05 * 0.95 0.91 0.93 1.11
39,290 21,140 28,040 29,560
KCl
N/m 37490 36930 37810 35610 30,360 39,460 39,410 7800 31,670 29,500 29,860 42,970 17,540 32,400 30,480
NaCl
As
N/m
As
1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
41590 41350 41580 43440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
1.12 1.42 0.93 0.95 0.98 1.01 1.04 1.40 1.66 1.28 0.97 0.95 0.87 0.99 0.92 0.95
N/m 37300 37000 37560 37330 29810 40,130 37,660 6120 29,800 30,320 41,750 41,860 43,520 29,950 31,970 38,380
LiCl As 1.17 1.34 0.94 0.96 0.95 0.99 1.02 1.19 1.62 1.41 1.11 0.93 1.00 0.93 0.92 1.19
N/m 37,870 34,850 38,610 37,360 30,540 40,490 39,540 4970 30,670 32,550 30,690 41,020 43,150 29,440 33,560 30,450
Comparison between SCX and C18 columns The results obtained on the SCX column were compared with those obtained on a C18 stationary phase. The retention and separation selectivity can be compared using tR1 vs tR2 correlations (Fig. 1). The tR values were obtained in optimal eluent systems: on a C18 column with mobile phase containing 25% MeCN, acetate buffer at pH 3.5, 0.025 M DEA and on a SCX column with mobile phase containing 25% MeCN, phosphate buffer at pH 2.5 and 100 mM KCl. The correlation points are dispersed, which indicates great differences in selectivity on SCX and C18 stationary phases.
1704
Figure 1. Correlation of retention time (tR) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at pH 3.5; 0.025 mol –1 per liter ML diethylamine (DEA). SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
wileyonlinelibrary.com/journal/bmc
Figure 2. Chromatograms obtained for mixture of standards (for symbols see Table 2) obtained on: (A) C18 column in eleunt system containing 35% MeCN, buffer at pH 3.5 (200 mM CH3COOH and 200 mM CH3COONa); and (B) SCX column in eleunt system containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs For example, fluoxetine and medazepam or alprazolam, imipramine and hydroxizine eluted in a narrow range on a C18 column are sufficiently well separated on the SCX column. There are, however, groups of psychotropic drugs better separated on C18 stationary phase, for example, escitalopram, imipramine, biperiden, sertraline and rivastigmine or sulpiride, medazepam and sertindole. Fig. 2 presents chromatograms of standard mixtures separated by the use of both columns. All mixtures can be separated neither in RP nor in IEC systems. However, more separated individual peaks can be observed on ion-exchange column in similar time.
Figure 3 presents the comparison of As values obtained for investigated drugs on C18 and SCX columns. A comparison of the data shows that better symmetry of the peaks was obtained for most investigated compounds on the SCX stationary phase; only for rivastigmine and medazepam were peaks more symmetrical on the C18 column. For other compounds more symmetrical peaks on the SCX column were obtained. On the SCX column only for medazepam was As not in the acceptable range, but on C18 six drug peaks were asymmetrical. In particular, a great improvement in the symmetry of the peaks compared with the SCX column with a C18 was obtained for escitalopram, sertindol, imipramine and fluoxetine for which on the C18 column peaks were very asymmetrical. Higher values of N/m were obtained for eight psychotropic drugs on the SCX column in comparison with those obtained on the C18 column, but only for alprazolam on C18 column were N/m values signifivantly higher than on SCX column; for rivastigmine and medazepam N/m values obtained on C18 stationary phase were slightly higher than N/m values obtained on the SCX colum (on both columns N/m > 40,000; Fig. 4).
Validation of the analysis of escitalopram, sulpiride and zolpidem in serum
Figure 3. Correlation of asymmetry factor (As) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at pH 3.5; 0.025 –1 ML DEA. SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl. Lines parallel to x- and y-axes restrict correct As values in both systems.
On the basis of the results of chromatographic system optimization, an optimal system for the analysis of some psychotropic drugs in human serum was selected. Mixtures being separated and determined in human serum are often used in multidrug therapy. Most assays for determination of basic psychotropic drugs have been conducted on different C18 columns although often asymmetrical peaks and low systems efficiency are obtained on these columns. In our experiments on the XBridge C18 column As values of escitalopram and zolpidem were 1.81 and 2.20, respectively, and for sulpiride the peak was very asymmetrical and fuzzy. On the SCX column the following As values were obtained: 0.95 for escitalopram; 1.48 for zolpidem; and 0.99 for sulpiride. System efficiency was also higher on SCX column than on the C18 column for all quantified drugs (N/m was 41,580 for escitalopram, 34,890 for zolpidem and 44,240 for sulpiride) on SCX column, while on the
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
1705
Figure 4. Graphical comparison of theoretical plate number (N/m) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at –1 pH 3.5; 0.025 ML DEA. SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
A. Petruczynik et al.
Figure 5. Chromatogram obtained for human serum sample fortified with: escitalopram – E; sulpiride – S; and zolpidem – Z. Chromatograms were obtained on an SCX column in an eleunt system containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
Table 5. Parameters of calibration curve for selected psychotropic drugs Name of compounds Escitalopram Sulpiride Zolpidem
Range (μg/mL)
Equation of calibration curve
r
LOD
LOQ
0.1–10 0.1–10 0.1–10
y = 140,312x + 14,603 y = 176,726x + 12,889 y = 309,112x – 71,102
0.9997 0.9998 0.9998
0.2837 0.2166 0.2270
0.8597 0.6563 0.6878
C18 column for escitalopram N/m was 13,710, for zolpidem it was 16,950 and for sulpiride a fuzzy peak was obtained. All calibration curves were linear over the concentration ranges with correlation coefficients (r) >0.9996. The chromatogram of a human serum fortified by three psychotropic drugs is presented in Fig. 5. Before HPLC analysis fortified human serum samples were prepared by the procedure described in the Experimental section. The proposed method was validated by linearity, limit of detection (LOD), limit of quantification (LOQ) and precision (Table 5). The identities of analyte peaks in human serum samples were confirmed by comparison of their UV spectra with the spectra of standards. The following recoveries were determined: 98.6% for escitalopram; 71.9% for sulpiride; and 89.7% for zolpidem.
Conclusions
1706
Ion-exchange resins can be applied in determination of basic drugs. The best results were obtained when different chlorides were added to mobile phase containing organic modifier and phosphate buffer. The application of eluents containing various organic modifiers (MeOH or MeCN) with different cations in salts used to prepare the phosphate buffers and different chlorides caused significantly differences in retention, separation selectivity, system efficiency and peak symmetry in ion-exchange systems. The best efficiency and most symmetrical peaks in case of the investigated drugs were observed for mobile phases with potassium chloride and potassium phosphate as buffer components and MeCN as organic modifier. A comparison of the data obtained on SCX column with those obtained on C18 stationary phase shows that better symmetry of
wileyonlinelibrary.com/journal/bmc
the peaks was obtained for almost all investigated compounds on the SCX stationary phase. Similarly higher efficiency was also obtained for most investigated psychotropic drugs on the SCX column in comparison with those obtained on the C18 column. Both systems were selective for separation of the analytes. The most selective and efficient mobile phase system on SCX column was successfully applied for separation and quantification of selected psychotropic drugs in fortified human serum samples.
References Brown J, Chighine A, Colucci MA, Galaffu N, Hirst SC, Seymour HM, Shiers JJ, Wilkes RD, Williams JG and Wilson JRH. New functionalised silicas for highly selective cation exchange SPE purification in medicinal chemistry. Tetrahedron Letters 2008; 49: 4968–4971. Croes K, McCarthy PT and Flanagan RJ. HPLC of basic drugs and quaternary ammonium compounds on microparticulate strong cationexchange materials using methanolic or aqueous methanol eluents containing an ionic modifier. Journal of Chromatography A 1995; 693: 289–306. Ghosheh OA, Browne D, Rogers T, de Leon J, Dwoskin LP and Crooks PA. A simple high performance liquid chromatographic method for the quantification of total cotinine, total 3%-hydroxycotinine and caffeine in the plasma of smokers. Journal of Pharmaceutical and Biomedical Analysis 2000; 23: 543–549. Gilart N, Cormack PAG, Marcé RM, Fontanals N and Borrull F. Selective determination of pharmaceuticals and illicit drugs inwastewaters using a novel strong cation-exchange solid-phaseextraction combined with liquid chromatography–tandem massspectrometry. Journal of Chromatography A 2014; 1325: 137–146. Jiang H, Zhang Y, Ida M, LaFayette A and Fast DM. Determination of carboplatin in human plasma using HybridSPE-precipitation along with liquid chromatography–tandem mass spectrometry. Journal of Chromatography B 2011; 879: 2162–2170.
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs Kawano S, Takahashi M, Hine T, Yamamoto E and Asakawa N. On-line pretreatment using methylcellulose-immobilized cation-exchange restricted access media for direct liquid chromatography/mass spectrometric determination of basic drugs in plasma. Rapid Communications in Mass Spectrometry 2005; 19: 2827–2832. Koseki N, Kawashita H, Niina M, Nagae Y and Masuda N. Development and validation for high selective quantitative determinationof metformin in human plasma by cation exchanging with normal-phase LC/MS/MS. Journal of Pharmaceutical and Biomedical Analyis 2005; 36: 1063–1072. Laussmann T and Meier-Giebing S. Forensic analysis of hallucinogenic mushrooms and khat (Catha edulis FORSK) using cation-exchange liquid chromatography. Forensic Science International 2010; 195: 160–164. Law B and Appleby JRG. Re-evaluation of strong cation-exchange highperformance liquid chromatography for the analysis of basic drugs. Journal of Chromatography A 1996; 725: 335–341. Law B and Appleby JRG. The application of strong cation exchange highperformance liquid chromatography to drug analysis. Journal of Pharmaceutical and Biomedical Analyis 1998; 17: 1199–1203. Li-Boa D, Rong-Hua Z, Huan-Dea L, Feng W, Ping-Fei F and Jiang L. Quantitative analysis of trazodone in human plasma by using HPLCfluorescence detector coupled with strong cation exchange chromatographic column: application to a pharmacokinetic study in Chinese healthy volunteers. Journal of Chromatography B 2014; 944: 43–48. Long Z, Wang C, Guo Z, Zhang X, Nordahl L and Liang X. Strong cation exchange column allow for symmetrical peak shape and increased sample loading in the separation of basic compounds. Journal of Chromatography A 2012; 1256: 67–71. Long Z, Guo Z, Xue X, Zhang X and Liang X. Two-dimensional strong cation exchange/positively charged reversed-phase liquid chromatography for alkaloid analysis and purification. Journal of Separation Science 2013a; 36: 3845–3852. Long Z, Guo Z, Xue X, Zhang X, Nordahl L and Liang X. Selective separation and purification of highly polar basic compoundsusing a silica-based strong cation exchange stationary phase. Analytica Chimica Acta 2013b; 804: 304–312. McCalley DV. The challenges of the analysis of basic compounds by high performance liquid chromatography: Some possible approaches for improved separations. Journal of Chromatography A 2010; 1217: 858–880.
Morgan PE, Tapper J and Spencer EP. Measurement of total mirtazapine and normirtazapine in plasma/serum by liquid chromatography with fluorescence detection. Journal of Chromatography B 2003; 798: 211–215. Nawrocki J. The silanol group and its role in liquid chromatography. Journal of Chromatography A 1997; 779: 29–71. Neubauer K. Advantages and disadvantages of different column types for speciacion analysis by LC-ICP-MS. Spectroscopy 2009; 24: 30–32. www.spectrosopyonline.com Newby JJ, Legg MA, Rogers B and Wirth MJ. Annealing of silica to reduce the concentration of isolated silanols and peak tailing in reverse phase liquid chromatography. Journal of Chromatography A 2011; 1218: 5131–5135. Niemann RA and Gay ML. Determination of ephedrine alkaloids and synephrine in dietary supplements by column-switching cation exchange high-performance liquid chromatography with scanningwavelength ultraviolet and fluorescence detection. Journal of Agriculture and Food Chemistry 2003; 51: 5630–5638. Nikitas P and Pappa-Louisi A. Retention models for isocratic and gradient elution in reversed-phase liquid chromatography. Journal of Chromatography A 2009; 1216: 1737–1755. Petruczynik A and Waksmundzka-Hajnos M. High performance liquid chromatography of selected alkaloids inion-exchange systems. Journal of Chromatography A 2013; 1311: 48–54. Plenis A and Bączek T. Modern chromatographic and electrophoretic measurements of antidepressants and their metabolites in biofluids. Biomedical Chromatography 2011; 25: 164–198. Schonberg L, Grobosch T, Lampe D and Kloft C. New screening method for basic compounds in urine by on-line extraction–high-performance liquid chromatography with photodiode-array detection. Journal of Chromatography A 2006; 1134: 177–185. Snyder LR, Kirkland JJ and Glajch JL. Practical HPLC Method Development. Wiley: New York , 1997. Walshe M, Kelly MT, Smyth MR and Ritchie H. Retention studies on mixedmode columns in high-performance liquid chromatography. Journal of Chromatography A 1995; 708: 31–40. Walshe M, Kelly MT and Smyth MR. Comparison of two extraction methods for determination of propranolol and furosemide in human plasma by mixed-mode chromatography. Journal of Pharmaceutical and Biomedical Analyis 1996; 14: 475–481.
1707
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
Revised: 24 February 2015,
Accepted: 30 March 2015
Published online in Wiley Online Library: 5 May 2015
(wileyonlinelibrary.com) DOI 10.1002/bmc.3482
Ion-exchange vs reversed-phase chromatography for separation and determination of basic psychotropic drugs Anna Petruczynik*, Karol Wróblewski, Michał Deja and Monika Waksmundzka-Hajnos ABSTRACT: Ion exchange chromatography, an alternative to reversed-phase (RP) chromatography, is described in this paper. We aimed to obtain optimal conditions for the separation of basic drugs because silica-based RP stationary phases show silanol effect and make the analysis of basic analytes hardly possible. The retention, separation selectivity, symmetry of peaks and system efficiency were examined in different eluent systems containing different types of buffers at acidic pH and with the addition of organic modifiers: methanol and acetonitrile. The obtained results reveal a large influence of the salt cation used for buffer preparation and the type of organic modifier on the retention behavior of the analytes. These results were also compared with those obtained on an XBridge C18 column. The obtained results demonstrated that SCX stationary phases can be successfully used as alternatives to C18 stationary phases in the separation of basic compounds. The most selective and efficient chromatographic systems were applied for the quantification of some psychotropic drugs in fortified human serum samples. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: IEC-HPLC; RP-HPLC; psychotropic drugs; optimization; quantitative analysis
Introduction
1700
Depression is one of the most frequently occurring psychiatric conditions affecting the economic and social functioning of people all over the world. Therapeutic drug monitoring of antidepressants provides the possibility of reducing side effects and optimizing the treatment of patients with depression. For this reason, there is a need to develop analytical methods for the analysis of these drugs, especially for their determination in biological fluids (Plenis and Bączek, 2011). A wide variety of analytical techniques has been used to determine different psychotropic drugs. Nowadays liquid chromatography in reversed-phase system (RP HPLC) is the most frequently used method (Nikitas and Pappa-Louisi, 2009). However, the separation of basic compounds in LC-RP systems still remains problematic owing to their interactions with free-residue silanol groups. Organic bases appear in aqueous solution as ionized and unionized forms that interact differently with the stationary phase surface-active sites (ion-exchange, hydrophobic and H-bond interaction; McCalley, 2010). The peak shape of basic compounds depends on the kinetics of the different interactions. In aqueous mobile phases, interactions of ions are usually stronger and slower than hydrophobic interactions. This situation leads to peak tailing, low efficiency and poor column-to-column reproducibility (Newby et al., 2011). Interactions with the silanols can be reduced by the use of mobile phases buffered at low pH, when silanol ionization is suppressed, or at high pH, to suppress solute ionization, addition of anionic ion-pair reagents to form neutral associates, addition of organic amines as silanol blockers, or the use of an appriopriate stationary phase (Nawrocki, 1997). Although advances in column technology, especially in the use of high-purity silica, have given considerable improvements for the analysis of bases, low
Biomed. Chromatogr. 2015; 29: 1700–1707
efficiency and tailing peaks continue to be problematic for these compounds under some circumstances. Therefore, an alternative approach to modulate the retention, peak shape, systems efficiency and separation selectivity of ionizable compounds using of ion exchange chromatography (IEC) is required (Long et al., 2012). The application of IEC allows high efficiency, symmetrical peaks and usually good separation to be achieved using relatively simple eluents containing only buffer or a mixture of buffer and organic modifier (Long et al., 2013b). Additionally, mobile phases applied in IEC contain usually low concentrations of organic modifiers (or do not contain any), which makes the method environmentally friendly. In eluent systems at constant pH, the retention of basic compounds in the chromatographic mode is inversely proportional to ionic strength and (if addition of organic modifier is used) to kind and concentration of organic modifier in mobile phases (Croes et al., 1995). Because of the similarity of ion-exchange and ion-pair HPLC retention, many separations that are possible using IEC can also be achieved using ion-pairing chromatography (IPC). For the separation of typical small-molecule samples, IPC may have certain advantages: higher column efficiencies, easier control over selectivity and resolution, and more stable and reproducible columns (Snyder et al., 1997). However, both IEC and reversed-phase
* Correspondence to: A. Petruczynik, Department of Inorganic Chemistry, Medical University of Lublin, Chodzki 4a, Lublin 20-093, Poland. Email: [email protected] Department of Inorganic Chemistry, Medical University of Lublin, Chodzki 4a, Lublin 20-093, Poland Abbreviations used: DEA, diethylamine; IEC, ion exchange chromatography.
Copyright © 2015 John Wiley & Sons, Ltd.
Ion-exchange chromatography for analysis of psychotropic drugs
Biomed. Chromatogr. 2015; 29: 1700–1707
Mixed mode C18/SCX phases also give good peak shape for ionized bases, e.g. propranolol and furosemide were determined in human plasma (Walshe et al., 1996). The analysis of drugs in biological samples such as plasma, urine and saliva requires robust, precise and fast methods. The success of such analyses is largely dependent not only on the choice of the most optimal chromatographic system but also on the quality of the sample preparation. Different SCX stationary phases for SPE were also used for preparation of biological samples containing basic analytes (Brown et al., 2008; Gilart et al., 2014). The aim of this paper was to investigate the various parameters influencing the retention, peak shape, efficiency and separation selectivity of some psychotropic drugs on an SCX column. The effect of the use of MeOH or MeCN as organic modifiers, ionic strength and type of buffer applied in the mobile phase was examined. The results obtained on the SCX column were compared with those obtained on C18 stationary phase. Systems with the best parameters and highest selectivity were applied for the quantification of some psychotropic drugs in fortified samples of human plasma.
Experimental The analysis was performed using liquid Shimadzu chromatograph LC-10 ATVP equipped with a Luna SCX 150 × 4.6 mm, 5 μm particle size column (Phenomenex, USA) or an XBridge C18 column from Waters (150 × 4.6 mm, 5 μm). The properties of the columns are presented in Table 1. The chromatograph was equipped with a Shimadzu detector SPD–10 AVVP and a Rheodyne 20 μL injector. The detection was carried out at 254 nm wavelength. All chromatographic measurements were carried out at 22°C controlled by a CTO-10ASVP thermostat. Eluent flow rate was 1.0 mL/min. The column pressure during the analysis was in the range of 250–300 psi. MeCN and MeOH of chromatographic quality were from E. Merck (Darmstadt, Germany) and diethylamine (DEA) (analytical reagent grade) was from Sigma-Aldrich. Potassium dihydrogen phosphate, sodium dihydrogen phosphate, calcium dihydrogen phosphate, ammonium dihydrogen phosphate, phosphoric acid (85%), sodium chloride, potassium chloride, ammonium chloride, cesium chloride, lithium chloride, barium chloride, calcium chloride, ammonium acetate and acetic acid were of p.a. quality (Polish Reagents Gliwice, Poland). The pH of buffers used in experiments was measured using a pH meter CP-505 (Elmetron, Zabrze, Poland) in aqueous solutions. Analysis of serum samples was performed using a liquid chromatograph LaChrom Elite (Merck) equipped with autosampler diode array detector (DAD). All chromatographic measurements were carried out at 22°C with an eluent flow rate of 1.0 mL/min using the same IEC column. The DAD was set in the 200–400 nm range and quantitative analysis was performed at 238 nm for all compounds. The chromatographic data was acquired and processed by LaChrom Elite HPLC software (Merck). All chromatographic parameters such as retention times (tR), asymmetry factor (As) (calculated as 10% of peak height) and theoretical plate number (N/m) were calculated by software CLASS-VP 5.0 controlling the chromatograph.
Table 1. Physicochemical properties of columns Column
Particle Pore Surface size (μm) size (Å) area (m2/g)
Xbridge C18 SCX
Copyright © 2015 John Wiley & Sons, Ltd.
5 5
130 100
Carbon load
pH limits
185 18% 1–12 400 0.55% sulfur load 2–7
wileyonlinelibrary.com/journal/bmc
1701
IPC have their advantages and disadvantages. One of the main advantages of IEC is that there is only one interaction involved in the separation: the analytical species interact with the stationary phase whereas in IPC two interactions are involved with the separation – the ion-pairing reagent interacting with stationary phase of the column and the species interacting with the ion-pairing reagent. As a result, IEC may have more matrix tolerance. The disadvantage of IEC is that these columns typically are much more expensive than reversed-phase columns and ion-exchange columns may not always be well established, which could lead to poor reproducibility from column to column (Neubauer, 2009). The effects of changes in eluent composition, kind of buffer, ionic strength and pH on the retention of different basic compounds on C18/strong cation-exchange (SCX) columns in eluent systems containing acetonitrile or methanol and sodium, potassium or ammonium phosphate in different concentrations and at different pH values have been investigated by some authors (Walshe et al., 1995). An SCX column was applied to the analysis of various basic drugs, e.g. β-blockers, using a mobile phase containing methanol (MeOH) and ammonium formate with trifluoroacetic acid at pH 2.45 (Law and Appleby, 1996), trazodone in a mobile phase system containing acetonitrile (MeCN) and ammonium phosphate adjusted to pH 6.0 (Li-Boa et al., 2014), MeOH with sodium phosphate buffer adjusted to pH 4.8 with triethylamine (Ghosheh et al., 2000), MeOH and ammonium perchlorate (Morgan et al., 2003); often mixtures of MeCN and ammonium formate or acetate as additives to mobile phase were used ( Jiang et al., 2011; Koseki et al., 2005). Ephedrine alkaloids in dietary supplements were determined on SCX column with mobile phase containing MeCN, sodium phosphate buffer at pH 3.0 (Niemann and Gay, 2003) and alkaloids from hallucinogenic mushrooms with mobile phase containing EtOH and phosphate buffer with the addition of NaCl (Laussmann and Meier-Giebing, 2010). Ion-exchange chromatography has been used successfully to separate different basic compounds in biological samples, e.g. alkaloids in urine (Schonberg et al., 2006), or in plasma (Ghosheh et al., 2000), some basic drugs in plasma (Law and Appleby, 1998), the antidepressant drug trazodone in human plasma ( Jiang et al., 2011), mitrazapine and its metabolite in plasma (Morgan et al., 2003) and the antihyperglycemic drug metformin in human plasma (Koseki et al., 2005). Some tricyclic antidepressants were determined in plasma samples on methylcellulose-immobilized restricted access media column with strong cation-exchange groups on an internal surface (MC-SCX; Kawano et al., 2005). Long et al. (2012) described the results of the influence of salt concentration and buffer pH on the peak shape obtained for some basic compounds under high sample loading conditions. Basic compounds were effectively separated on an SCX column and only a small increase in peak width was observed as the loading amount of analytes increased, which can be important for preparative separation of these substances. The influence of type of buffer at different values of pH, ionic strength, type and concentration of organic modifier on the retention of selected alkaloids on SCX column was also investigated (Petruczynik and Waksmundzka-Hajnos, 2013). The SCX column is also used coupled with an RP column for multidimensional separation owing to the good orthogonality between SCX and RP separation mechanisms with similar mobile phases. For example, a 2D-LC system, SCX/RP, has been developed to eliminate nonalkaloids and to achieve symmetrical peak shape and high orthogonality selectivity for alkaloid separation (Long et al., 2013a).
A. Petruczynik et al. Extraction procedure Solid-phase extraction (SPE) was carried out using Bakerbond SPE C18 endcaped columns and an SPE chamber–Baker SPE–12G ( J. T. Baker, Philipsburg, PA, USA). The SPE method was optimized and the best procedure in terms of recovery and purification was selected for sample preparation. A 1 mL aliquot of human serum sample was fortified by escitalopram, sulpiride and zolpidem at concentration 0.5 μg/mL and was incubated at 37°C for 60 min. A C18 endcaped cartridge was preactivated by passing through 3 mL of MeCN, 10 mL of MEOH and 30 mL of water. A 1 mL aliquot of serum was mixed with 16 mL of water and 3 mL of 1 M sodium carbonate. The sample preparation was mixed and loaded onto the C18 cartridge at a flow-rate of 2 mL/min. The cartridge was washed twice with 5 mL of water. Finally, the drugs were eluted with 5 mL (two fractions, 3 + 2 mL) of acetonitrile and HCOOH (80%; 19:1). The sample was evaporated to dryness, the residue was dissolved in 0.5 mL of acetonitrile and a volume of 20 μL was injected into chromatographic systems.
Validation protocol The proposed method was validated by linearity, limit of detection (LOD), limit of quantification (LOQ) and precision . Method linearity was studied by analyzing solvent-based standard solutions in triplicate at eight concentrations ranging from 0.1 to 10 μg/mL for escitalopram, sulpiride and zolpidem. The LOD and LOQ were calculated according to the formulas LOD = 3.3(SD/S) and LOQ = 10(SD/S), respectively, where SD is the standard deviation of the response and S is the slope of the calibration curve. aThe accuracy of the method was tested by performing recovery studies. The average recovery was 98.6% for escitalopram, 89.66% for zolpidem and 71.86% for sulpiride.
Results and discussion Psychotropic drug standards (Table 2) were chromatographed on an SCX column in eluent systems containing different buffers at pH 2.5 consisting of various chlorides and organic modifiers or
on a C18 column in eluent systems containing acetonitrile, phosphate buffer at pH 3.5 and 0.025 M DEA. These chromatographic systems were compared in terms of retention of investigated drugs, differences in selectivity, peak shape and performance. The use of different buffers in mobile phase – acetate, formate and phosphate without addition of chlorides – were tested, but asymmetrical peaks and low systems efficiency were obtained.
Selection of the organic modifier The first part of our work was a study of the retention behavior of investigated drugs on an SCX column in eluent systems containing phosphate buffer at pH 2.5, KCl and MeOH or MeCN as organic modifiers. Great differences in separation selectivity were observed between systems with mobile phases containing different organic modifiers (Table 2). The retention of psychotropic drugs was stronger in mobile phase with MeOH. Also, differences in peak shape and system efficiency were obtained in eluent systems containing MeOH or MeCN. Symmetry of peaks was significantly improved in mobile phase containing MeCN. In systems with MeOH as organic modifier, only in one out of 16 psychotropic drugs was the value of As in the acceptable range (0.8–1.5), while in systems with mobile phase containing MeCN, for 15 drugs the symmetry of peaks was acceptable and for 12 of them As values were excelent (0.9 < As < 1.2). The addition of MeCN to mobile phases also caused an increase in systems efficiency. In systems containing MeOH, N/m values were >25,000 for five out of 16 psychotropic drugs, and in systems containing MeCN as organic modifier in the mobile phase, for 15 investigated drugs, the number of theoretical plates (N/m) was >25,000. Because of the low efficiency and high peak assymmetry, the resolution of investigated drugs in systems with methanol got worse and much better results were observed when acetonitrile as modifier was applied. For this reason in further experiments MeCN was used as organic modifier.
Table 2. Values of retention time (tR), asymmetry factor (As) and theoretical plate number (N/m) for investigated psychotropic drugs obtained on an SCX column in eluent systems containing 25% organic modifier, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl Investigated psychotropic drugs
Abbreviation
Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moclobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
R M E T Sd I Ro A Z H Mo B F Se Fl S
MeOH
MeCN
tR
As
N/m
tR
As
N/m
17.44 31.84 22.21 10.78 65.13 26.83 19.43 * 51.61 81.27 15.33 15.67 16.03 22.21 12.43 20.83
3.63 4.56 2.89 1.55 4.60 3.55 2.85
18,470 9910 23,070 31,660 40,030 14,230 22,160
4.21 2.74 2.30 1.94 1.42 2.99 2.00 1.54
8870 10,890 30,190 38,500 35,720 14,960 14,620 33,130
8.18 11.03 7.47 6.21 10.23 8.21 8.50 4.30 13.22 19.70 8.39 8.34 8.34 7.54 5.37 9.78
1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
41,590 41,350 41,580 43,440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
1702
* Fuzzy peak.
wileyonlinelibrary.com/journal/bmc
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs Selection of buffer cation The next stage of the experiments concerned the effect of different cations in salts used to prepare the phosphate buffers as mobile phase additives. The use of different salts caused differences in retention, separation selectivity, system efficiency and peak symmetry. The selectivity and sequence of elution for the psychotropic drugs in the systems containing buffers with various cations were different (Table 3). For most drugs better selectivity was obtained in systems containing sodium salts. Also, differences in peak shape and system efficiency were obtained in eluent systems containing salts with different cations. The worst peak shapes were observed in eluent containing sodium salts. Only nine out of 16 investigated drugs had As values in the acceptable range, whereas in the system with calcium cations As values for 12 compounds were acceptable. The most symmetrical peaks were obtained in systems containing potassium salts – for 15 psychotropic drugs As values were acceptable and for 12 they were excellent. High efficiency was also obtained in systems containing potassium salts in the mobile phase: for 15 drugs, N/m was >25,000, while in systems containing calcium or sodium salts for 9 and 12 compounds N/m was >25,000, respectively. For this reason further investigation was performed in eluent systems containing potassium phosphate buffer.
Selection of the chloride salt cation Because the retention of the investigated psychotropic drugs significantly depends on the type of cation used to prepare buffer solutions, mobile phases containing potassium phosphate buffer and different chlorides were used in the next set of experiments. The retention and separation selectivity obtained using buffers containing chlorides with various cations (pH and the other eluent components are the same) were significantly different (Table 4).
The eluent strength decreased from the eluent containing barium chloride to the eluent containing lithium chloride according to Hofmeister series for most investigated drugs. Displacer strength in cation-exchange chromatography highly depends on the valency of cations. Thus the displacer strength of bivalent cations is higher than that of monovalent and trivalent ones (Snyder et al., 1997). Because of that, in our investigations analysed compounds were eluted from the column after a very short time and they were poorly separated in systems containing chlorides of bivalent cations (calcium and barium). In mobile phase containing cesium chloride, separation of most drugs was also poor. The similar retention times and good separation selectivities were obtained in eluents containing additions of ammonium, potassium and sodium chlorides. Various chlorides cause also differences in peak symmetry, e.g. for alprazolam As = 2.66 in the system with CaCl2, whereas in the system with LiCl, As = 1.19; for flupentixole As values were 0.98 in systems with CsCl and KCl and 1.00 in systems with added LiCl, but in the system with BaCl2 the peak was very asymmetrical. However, for most investigated psychotropic drugs in systems with potassium phosphate and different chlorides, peaks were symmetrical; only in a few cases As values dif not posess the optimum range. The differences were also observed for system efficiency, for example, for rivastigmine N/m was 18,390 for the system with CaCl2 and 41,590 for the system with KCl; for flupentixol N/m was about 19,000 for the system with BaCl2 and N/m was >40,000 for systems with NH4Cl, KCl, NaCl and LiCl. For nearly all investigated drugs, worse system efficiency was obtained in eluents containing chlorides with bivalent cations compared with eluent systems with monovalent cations. The most efficient system for most drugs was the system with addition of KCl to mobile phase – for 15 from 16 investigated drugs N/m was >25,000.
Table 3. Values of tR As and N/m for investigated psychotropic drugs obtained on an SCX column in eluent systems: (1) 25% MeCN, buffer at pH 2.5 [100 mM Ca(H2PO4)2, 100 mM H3PO4] and 100 mM CaCl2; (2) 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl; and (3) 25% MeCN, buffer at pH 2.5 (100 mM NaH2PO4, 100 mM H3PO4) and 100 mM NaCl Investigated 25% MeOH 100 mM CaCl2, 100 mM Ca (H2PO4)2, H3PO4, pH 2.5 psychotropic drugs As N/m tR Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moclobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
5.3 6.18 4.26 1.88 5.76 4.66 5.09 3.63 8.06 * 4.47 4.76 4.77 4.13 3.20 5.44
1.27 1.39 1.00 2.47 1.08 1.06 1.11 2.32 1.74
26,680 27,230 27,950 13,570 24,330 28,930 28,570 5530 21,970
1.05 1.00 0.84 1.08 1.04 0.97
30,640 30,970 30,460 20,010 22,300 30,110
25% MeOH 100 mM KCl, 100 mM KH2PO4, H3PO4, pH 2.5 tR 8.18 11.03 7.47 6.21 10.23 8.21 8.50 4.30 13.22 19.70 8.39 8.34 8.34 7.54 5.37 9.78
As 1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
N/m 41,590 41,350 41,580 43,440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
25% MeOH 100 mM NaCl, 100 mM NaH2PO4, H3PO4, pH 2.5 tR
As
13.87 12.04 7.36 9.69 14.48 21.78 9.83 * 24.58 * 13.05 8.85 12.61 6.69 6.82 18.82
N/m
2.60 1.46 0.91 1.01 0.75 2.03 1.09
19,730 39,220 38,240 37,520 18,380 31,130 38,590
2.07
27,350
1.80 0.93 1.12 0.97 0.90 1.15
39,310 44,240 28,960 30,640 34,520 39,270
1703
* Fuzzy peak.
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
A. Petruczynik et al. Table 4. Values of As and N/m for investigated psychotropic drugs obtained on an SCX column in eluent systems containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM of different chlorides Investigated psychotropic drugs
CaCl2
As
N/m
As
Rivastigmine Medazepam Escitalopram Tramadol Sertindol Imipramine Ropinirol Alprazolam Zolpidem Hydroxizine Moklobemide Biperiden Flupentixole Sertraline Fluoxetine Sulpiride
0.9 1.27 1.04 1.04 1.00 1.05 1.09 2.12 1.47 1.02 1.00 * 0.67 0.90 1.01 1.03
26,570 27,210 26,750 23,510 24,840 28,930 28,720 10,980 24,120 23,480 20,770
1.02 1.27 1.00 1.01 1.02 1.04 1.04 2.66 1.49 1.04 0.98 * 0.84 0.98 1.02 1.03
BaCl2
19,290 31,934 19,560 25,820
CsCl
N/m 18390 28600 28190 28040 24,520 30,340 29,550 7250 23,810 21,500 21,000 29,700 32,770 21,500 25,180
NH4Cl
As
N/m
As
1.06 1.22 0.92 0.98 0.95 0.96 0.99 1.47 1.34 0.94 0.96 * 0.98 0.89 0.99 1.01
33530 35150 33450 33060 28,990 35,830 36,700 11,760 30,750 29,560 27,470
1.11 1.34 0.91 0.96 0.95 0.98 1.00 1.30 1.52 1.26 1.05 * 0.95 0.91 0.93 1.11
39,290 21,140 28,040 29,560
KCl
N/m 37490 36930 37810 35610 30,360 39,460 39,410 7800 31,670 29,500 29,860 42,970 17,540 32,400 30,480
NaCl
As
N/m
As
1.16 1.63 0.95 0.99 1.10 1.11 1.05 1.38 1.48 1.50 1.02 1.00 0.98 1.10 1.04 0.99
41590 41350 41580 43440 33,630 44,300 43,440 7270 34,890 35,560 47,670 48,120 47,050 33,500 35,400 44,240
1.12 1.42 0.93 0.95 0.98 1.01 1.04 1.40 1.66 1.28 0.97 0.95 0.87 0.99 0.92 0.95
N/m 37300 37000 37560 37330 29810 40,130 37,660 6120 29,800 30,320 41,750 41,860 43,520 29,950 31,970 38,380
LiCl As 1.17 1.34 0.94 0.96 0.95 0.99 1.02 1.19 1.62 1.41 1.11 0.93 1.00 0.93 0.92 1.19
N/m 37,870 34,850 38,610 37,360 30,540 40,490 39,540 4970 30,670 32,550 30,690 41,020 43,150 29,440 33,560 30,450
Comparison between SCX and C18 columns The results obtained on the SCX column were compared with those obtained on a C18 stationary phase. The retention and separation selectivity can be compared using tR1 vs tR2 correlations (Fig. 1). The tR values were obtained in optimal eluent systems: on a C18 column with mobile phase containing 25% MeCN, acetate buffer at pH 3.5, 0.025 M DEA and on a SCX column with mobile phase containing 25% MeCN, phosphate buffer at pH 2.5 and 100 mM KCl. The correlation points are dispersed, which indicates great differences in selectivity on SCX and C18 stationary phases.
1704
Figure 1. Correlation of retention time (tR) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at pH 3.5; 0.025 mol –1 per liter ML diethylamine (DEA). SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
wileyonlinelibrary.com/journal/bmc
Figure 2. Chromatograms obtained for mixture of standards (for symbols see Table 2) obtained on: (A) C18 column in eleunt system containing 35% MeCN, buffer at pH 3.5 (200 mM CH3COOH and 200 mM CH3COONa); and (B) SCX column in eleunt system containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs For example, fluoxetine and medazepam or alprazolam, imipramine and hydroxizine eluted in a narrow range on a C18 column are sufficiently well separated on the SCX column. There are, however, groups of psychotropic drugs better separated on C18 stationary phase, for example, escitalopram, imipramine, biperiden, sertraline and rivastigmine or sulpiride, medazepam and sertindole. Fig. 2 presents chromatograms of standard mixtures separated by the use of both columns. All mixtures can be separated neither in RP nor in IEC systems. However, more separated individual peaks can be observed on ion-exchange column in similar time.
Figure 3 presents the comparison of As values obtained for investigated drugs on C18 and SCX columns. A comparison of the data shows that better symmetry of the peaks was obtained for most investigated compounds on the SCX stationary phase; only for rivastigmine and medazepam were peaks more symmetrical on the C18 column. For other compounds more symmetrical peaks on the SCX column were obtained. On the SCX column only for medazepam was As not in the acceptable range, but on C18 six drug peaks were asymmetrical. In particular, a great improvement in the symmetry of the peaks compared with the SCX column with a C18 was obtained for escitalopram, sertindol, imipramine and fluoxetine for which on the C18 column peaks were very asymmetrical. Higher values of N/m were obtained for eight psychotropic drugs on the SCX column in comparison with those obtained on the C18 column, but only for alprazolam on C18 column were N/m values signifivantly higher than on SCX column; for rivastigmine and medazepam N/m values obtained on C18 stationary phase were slightly higher than N/m values obtained on the SCX colum (on both columns N/m > 40,000; Fig. 4).
Validation of the analysis of escitalopram, sulpiride and zolpidem in serum
Figure 3. Correlation of asymmetry factor (As) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at pH 3.5; 0.025 –1 ML DEA. SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl. Lines parallel to x- and y-axes restrict correct As values in both systems.
On the basis of the results of chromatographic system optimization, an optimal system for the analysis of some psychotropic drugs in human serum was selected. Mixtures being separated and determined in human serum are often used in multidrug therapy. Most assays for determination of basic psychotropic drugs have been conducted on different C18 columns although often asymmetrical peaks and low systems efficiency are obtained on these columns. In our experiments on the XBridge C18 column As values of escitalopram and zolpidem were 1.81 and 2.20, respectively, and for sulpiride the peak was very asymmetrical and fuzzy. On the SCX column the following As values were obtained: 0.95 for escitalopram; 1.48 for zolpidem; and 0.99 for sulpiride. System efficiency was also higher on SCX column than on the C18 column for all quantified drugs (N/m was 41,580 for escitalopram, 34,890 for zolpidem and 44,240 for sulpiride) on SCX column, while on the
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
1705
Figure 4. Graphical comparison of theoretical plate number (N/m) values obtained in the systems: C18 column – eluent, 25% MeCN; 20% acetate buffer at –1 pH 3.5; 0.025 ML DEA. SCX column – eluent, 25% MeCN; buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
A. Petruczynik et al.
Figure 5. Chromatogram obtained for human serum sample fortified with: escitalopram – E; sulpiride – S; and zolpidem – Z. Chromatograms were obtained on an SCX column in an eleunt system containing 25% MeCN, buffer at pH 2.5 (100 mM KH2PO4, 100 mM H3PO4) and 100 mM KCl.
Table 5. Parameters of calibration curve for selected psychotropic drugs Name of compounds Escitalopram Sulpiride Zolpidem
Range (μg/mL)
Equation of calibration curve
r
LOD
LOQ
0.1–10 0.1–10 0.1–10
y = 140,312x + 14,603 y = 176,726x + 12,889 y = 309,112x – 71,102
0.9997 0.9998 0.9998
0.2837 0.2166 0.2270
0.8597 0.6563 0.6878
C18 column for escitalopram N/m was 13,710, for zolpidem it was 16,950 and for sulpiride a fuzzy peak was obtained. All calibration curves were linear over the concentration ranges with correlation coefficients (r) >0.9996. The chromatogram of a human serum fortified by three psychotropic drugs is presented in Fig. 5. Before HPLC analysis fortified human serum samples were prepared by the procedure described in the Experimental section. The proposed method was validated by linearity, limit of detection (LOD), limit of quantification (LOQ) and precision (Table 5). The identities of analyte peaks in human serum samples were confirmed by comparison of their UV spectra with the spectra of standards. The following recoveries were determined: 98.6% for escitalopram; 71.9% for sulpiride; and 89.7% for zolpidem.
Conclusions
1706
Ion-exchange resins can be applied in determination of basic drugs. The best results were obtained when different chlorides were added to mobile phase containing organic modifier and phosphate buffer. The application of eluents containing various organic modifiers (MeOH or MeCN) with different cations in salts used to prepare the phosphate buffers and different chlorides caused significantly differences in retention, separation selectivity, system efficiency and peak symmetry in ion-exchange systems. The best efficiency and most symmetrical peaks in case of the investigated drugs were observed for mobile phases with potassium chloride and potassium phosphate as buffer components and MeCN as organic modifier. A comparison of the data obtained on SCX column with those obtained on C18 stationary phase shows that better symmetry of
wileyonlinelibrary.com/journal/bmc
the peaks was obtained for almost all investigated compounds on the SCX stationary phase. Similarly higher efficiency was also obtained for most investigated psychotropic drugs on the SCX column in comparison with those obtained on the C18 column. Both systems were selective for separation of the analytes. The most selective and efficient mobile phase system on SCX column was successfully applied for separation and quantification of selected psychotropic drugs in fortified human serum samples.
References Brown J, Chighine A, Colucci MA, Galaffu N, Hirst SC, Seymour HM, Shiers JJ, Wilkes RD, Williams JG and Wilson JRH. New functionalised silicas for highly selective cation exchange SPE purification in medicinal chemistry. Tetrahedron Letters 2008; 49: 4968–4971. Croes K, McCarthy PT and Flanagan RJ. HPLC of basic drugs and quaternary ammonium compounds on microparticulate strong cationexchange materials using methanolic or aqueous methanol eluents containing an ionic modifier. Journal of Chromatography A 1995; 693: 289–306. Ghosheh OA, Browne D, Rogers T, de Leon J, Dwoskin LP and Crooks PA. A simple high performance liquid chromatographic method for the quantification of total cotinine, total 3%-hydroxycotinine and caffeine in the plasma of smokers. Journal of Pharmaceutical and Biomedical Analysis 2000; 23: 543–549. Gilart N, Cormack PAG, Marcé RM, Fontanals N and Borrull F. Selective determination of pharmaceuticals and illicit drugs inwastewaters using a novel strong cation-exchange solid-phaseextraction combined with liquid chromatography–tandem massspectrometry. Journal of Chromatography A 2014; 1325: 137–146. Jiang H, Zhang Y, Ida M, LaFayette A and Fast DM. Determination of carboplatin in human plasma using HybridSPE-precipitation along with liquid chromatography–tandem mass spectrometry. Journal of Chromatography B 2011; 879: 2162–2170.
Copyright © 2015 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2015; 29: 1700–1707
Ion-exchange chromatography for analysis of psychotropic drugs Kawano S, Takahashi M, Hine T, Yamamoto E and Asakawa N. On-line pretreatment using methylcellulose-immobilized cation-exchange restricted access media for direct liquid chromatography/mass spectrometric determination of basic drugs in plasma. Rapid Communications in Mass Spectrometry 2005; 19: 2827–2832. Koseki N, Kawashita H, Niina M, Nagae Y and Masuda N. Development and validation for high selective quantitative determinationof metformin in human plasma by cation exchanging with normal-phase LC/MS/MS. Journal of Pharmaceutical and Biomedical Analyis 2005; 36: 1063–1072. Laussmann T and Meier-Giebing S. Forensic analysis of hallucinogenic mushrooms and khat (Catha edulis FORSK) using cation-exchange liquid chromatography. Forensic Science International 2010; 195: 160–164. Law B and Appleby JRG. Re-evaluation of strong cation-exchange highperformance liquid chromatography for the analysis of basic drugs. Journal of Chromatography A 1996; 725: 335–341. Law B and Appleby JRG. The application of strong cation exchange highperformance liquid chromatography to drug analysis. Journal of Pharmaceutical and Biomedical Analyis 1998; 17: 1199–1203. Li-Boa D, Rong-Hua Z, Huan-Dea L, Feng W, Ping-Fei F and Jiang L. Quantitative analysis of trazodone in human plasma by using HPLCfluorescence detector coupled with strong cation exchange chromatographic column: application to a pharmacokinetic study in Chinese healthy volunteers. Journal of Chromatography B 2014; 944: 43–48. Long Z, Wang C, Guo Z, Zhang X, Nordahl L and Liang X. Strong cation exchange column allow for symmetrical peak shape and increased sample loading in the separation of basic compounds. Journal of Chromatography A 2012; 1256: 67–71. Long Z, Guo Z, Xue X, Zhang X and Liang X. Two-dimensional strong cation exchange/positively charged reversed-phase liquid chromatography for alkaloid analysis and purification. Journal of Separation Science 2013a; 36: 3845–3852. Long Z, Guo Z, Xue X, Zhang X, Nordahl L and Liang X. Selective separation and purification of highly polar basic compoundsusing a silica-based strong cation exchange stationary phase. Analytica Chimica Acta 2013b; 804: 304–312. McCalley DV. The challenges of the analysis of basic compounds by high performance liquid chromatography: Some possible approaches for improved separations. Journal of Chromatography A 2010; 1217: 858–880.
Morgan PE, Tapper J and Spencer EP. Measurement of total mirtazapine and normirtazapine in plasma/serum by liquid chromatography with fluorescence detection. Journal of Chromatography B 2003; 798: 211–215. Nawrocki J. The silanol group and its role in liquid chromatography. Journal of Chromatography A 1997; 779: 29–71. Neubauer K. Advantages and disadvantages of different column types for speciacion analysis by LC-ICP-MS. Spectroscopy 2009; 24: 30–32. www.spectrosopyonline.com Newby JJ, Legg MA, Rogers B and Wirth MJ. Annealing of silica to reduce the concentration of isolated silanols and peak tailing in reverse phase liquid chromatography. Journal of Chromatography A 2011; 1218: 5131–5135. Niemann RA and Gay ML. Determination of ephedrine alkaloids and synephrine in dietary supplements by column-switching cation exchange high-performance liquid chromatography with scanningwavelength ultraviolet and fluorescence detection. Journal of Agriculture and Food Chemistry 2003; 51: 5630–5638. Nikitas P and Pappa-Louisi A. Retention models for isocratic and gradient elution in reversed-phase liquid chromatography. Journal of Chromatography A 2009; 1216: 1737–1755. Petruczynik A and Waksmundzka-Hajnos M. High performance liquid chromatography of selected alkaloids inion-exchange systems. Journal of Chromatography A 2013; 1311: 48–54. Plenis A and Bączek T. Modern chromatographic and electrophoretic measurements of antidepressants and their metabolites in biofluids. Biomedical Chromatography 2011; 25: 164–198. Schonberg L, Grobosch T, Lampe D and Kloft C. New screening method for basic compounds in urine by on-line extraction–high-performance liquid chromatography with photodiode-array detection. Journal of Chromatography A 2006; 1134: 177–185. Snyder LR, Kirkland JJ and Glajch JL. Practical HPLC Method Development. Wiley: New York , 1997. Walshe M, Kelly MT, Smyth MR and Ritchie H. Retention studies on mixedmode columns in high-performance liquid chromatography. Journal of Chromatography A 1995; 708: 31–40. Walshe M, Kelly MT and Smyth MR. Comparison of two extraction methods for determination of propranolol and furosemide in human plasma by mixed-mode chromatography. Journal of Pharmaceutical and Biomedical Analyis 1996; 14: 475–481.
1707
Biomed. Chromatogr. 2015; 29: 1700–1707
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
